Interactive HPC

Interactive HPC

What is Interactive HPC?

MSI’s interactive HPC resources are composed of some of the same systems used to support our batch scheduled HPC activities. The main difference between the two services is the way in which users interact with the MSI resource. Interactive HPC systems allow real-time user inputs in order to facilitate code development, real-time data exploration, and visualizations. Interactive HPC systems are used when data are too large to download to a desktop or laptop, software is difficult or impossible to install on a personal machine, or specialized hardware resources (e.g., GPUs) are needed to visualize large datasets. User input might come through a command line interface (i.e., system shell) while debugging a piece of software or through mouse clicks while using a program with a graphical interface.

What can I do with Interactive HPC?

Unlike MSI’s other computing options, the Interactive HPC resources are designed only to be used in real time for tasks such as interactive data exploration, creating plots or images, visualizations, or testing sections of code that will be used on other MSI systems.

Common use cases include:

  • Testing a compiled program at the command line before submitting it as a job
  • Developing code or exploring data in an interactive environment like MATLAB, IPython, or Rstudio
  • Visualizing data or building simulations in applications like Schrodinger, COMSOL, or Avizo
What types of Interactive HPC are available?
  • isub: Connects a user to an HPC node where they can use a command line or graphical interface to interactively run software.

  • NICE: Connects a user to a graphical environment, which may have specialized GPUs for visualizing large datasets.

    • Total Slots: 32 slots for accelerated remote desktops

    • Total Memory:  Up to 16 GB/session

    • Graphical Accelerators: Nvidia GK107GL

  • CITRIX: Citrix for Windows Environments: Connects a user to a graphical Windows environment, where they can access specialized Windows software for modeling and data analysis.

    • Nodes: 6 nodes for Windows applications

    • Total Memory: 128/node

    • Graphical Accelerators:  NVIDIA GRID K2

  • NX NoMachine: Connects a user to a graphical environment, which may be used as a lightweight remote desktop for connecting to other MSI resources.
  • Jupyter Notebooks: A web portal for notebook-based computing in the browser, which can be used for reproducible and shareable data analysis, visualization, and scripted control of larger tasks. Currently supports Python 2, Python 3, and R. This is an MSI Beta service.

 

Xen Citrix provides virtual Windows desktops with remote accessibility

Gateway to MSI’s interactive computing options

Method to display data sets for analysis.